Introduction
Heart failure (HF) is one of the major medical burdens in our society. Complete recovery of the heart after insult has been an attractive goal for many clinicians and scientists for long time. Mechanical circulatory support (MCS) devices, such as left ventricular assist devices (LVADs), effectively unload the heart by reducing the pressure overload/increased afterload, ultimately resulting in a degree of myocardial reverse remodeling. Currently, the predominantly used LVADs are continuous-flow devices, for patients suffering from end-stage HF.
The important research advantages of the VAD population are that they present us with a translational research medium through which investigations for new anti-remodeling and regenerative therapies for HF can be studied. Particularly, tissue specimens obtained from bridge to transplant (BTT) patients before LVAD implantation (referred to as “pre”) and at the time of LVAD explantation (referred to as “post”) provide a platform to perform molecular, structural, and functional studies. This in turn could pave the way for understanding the mechanisms behind progressive myocardial dysfunction during HF and the pathways that regulate reverse cardiac remodeling.
A subset of patients who undergo LVAD-induced mechanical unloading show significant improvements in cardiac structure and function to the point where some of these patients can be weaned from the support and achieve sustained recovery (i.e., responders), while the others show no such sign of recovery (i.e., nonresponders). Comparing responders and nonresponders in several aspects such as cardiac hypertrophy regression, contractile dysfunction, metabolism, cell death, extracellular matrix remodeling, and others could be helpful in revealing signatures of myocardial recovery, which in turn could help design novel HF therapies. However, in order to unmask the mechanisms driving myocardial recovery, the fundamental biological effects on myocardial structure and function must be understood. This chapter attempts to compile the ongoing clinical, translational, and basic science research in the field of MCS and cardiovascular remodeling.
Cardiac hypertrophy-atrophy
Myocardial unloading with a pulsatile flow LVAD has been shown to induce regression of cardiac myocyte hypertrophy. Animal models with prolonged unloading of nonfailing, nonhypertrophic myocardium (heterotopic transplantation, LVAD, or mitral regurgitation ) suggested that mechanical unloading could lead to cardiac myocyte atrophy. Based on these data it was postulated that this phenomenon may also apply to hypertrophic and failing myocardium. This notion of prolonged LVAD unloading in HF patients causing regression of cardiac hypertrophy to the point of atrophy prevailed until structural, ultrastructural, microstructural, metabolic, molecular, and clinical functional data indicated that prolonged continuous-flow LVAD unloading does not induce myocardial atrophy and degeneration ( Fig. 9.1 ). Specifically, in two human HF studies, unloading by means of pulsatile flow LVAD support decreased cardiac myocyte size, but not to the levels below respective normal cardiac myocytes. Furthermore, light ( Fig. 9.2 ) and confocal ( Fig. 9.3 ) microscopy findings complemented by ultrastructural, metabolic, and molecular data did not identify any evidence suggesting cardiac myocyte atrophy or degeneration during continuous-flow LVAD support. These data support echocardiographic observations in LVAD patients that show regression of hypertrophy without atrophy. Ongoing investigations have examined the roles of several complex regulatory pathways as critical components of these structural changes, including cyclooxygenase-2–induced Akt phosphorylation, mitogen-activated protein kinase/Erk, and Akt kinase/glycogen synthase kinase 3β. Also, a recent study reported that Akt activation stimulates glucose utilization but impairs mitochondrial oxidative capacity independently of cardiac hypertrophy. Whether the primary stimulus for the regression of hypertrophy is related directly to mechanical unloading/stretch or to circulating systemic factors needs further investigation.
Contractile dysfunction, calcium handling, and cytoskeletal proteins
Contractile function of the heart is conducted through composite processes such as sympathetic and parasympathetic heart rate control, core conduction system activities (thin [actin, tropomyosin, and troponin] and thick [myosin] filaments), calcium regulatory pathways that orchestrate shortening and relaxation, and humoral substances like catecholamines. The failing heart has high energy demands that place considerable stress on the cardiac contractile machinery, which in turn diminishes myocardial contractility. The myocyte contractile defects observed in failing hearts were shown to be reversed after pulsatile-flow LVAD unloading, with improved shortening and relaxation in isolated myocytes and isolated strips of ventricular tissue. Improvements in LVAD-induced contractile dysfunction can be partially explained by changes in Ca ++ handling, such as faster sarcolemmal Ca ++ entry and shorter action potential durations, higher sarcoplasmic reticulum Ca ++ content, improved abundance of sarcoplasmic reticulum (SR)/endoplasmic reticulum calcium ATPase, decreased abundance of Na + /Ca ++ exchanger, and beneficial changes in L-type calcium channel and ryanodine receptor (RyR) function. There is an increase in RyR clusters that are not near sarcolemmal L-type calcium channels due to reduced density of the tubular system (t-system); hence, they are not contained by couplons. This shift of RyR clusters away from sarcolemmal L-type calcium channels can affect the excitation-contraction coupling for two reasons: first, these nonjunctional RyR clusters can result in decreased calcium release from SR; and second, they create asynchronous calcium release due to a reduced open probability, which does not correlate with the quick opening of junctional RyR clusters. A recent study demonstrated a direct relationship linking the distance between nonjunctional RyR clusters and the change in left ventricular (LV) ejection fraction (EF) in post-LVAD patients. The authors reported that the EF improved when the distance was less than 1 μm, whereas a distance that exceeded 1 μm was not associated with improvement in EF. An intact t-system at the time of LVAD implantation may serve as a predictor for cardiac recovery induced through unloading ( Fig. 9.4 ).
The action potential initiates the excitation-contraction cycle in ventricular cardiomyocytes. Depolarization starts with sodium (Na + ) channels activating the L-type Ca ++ current that causes local Ca ++ -induced Ca ++ release (CICR) in the SR. The increased cytosolic Ca ++ binds to myofilaments, causing contraction followed by diastolic Ca ++ elimination, i.e., dissociation of Ca ++ from the contractile filaments and the excess cytosolic Ca ++ extruded via SR-Ca ++ ATPase, Na + -Ca ++ exchanger, and sarcolemmal Ca ++ pump. Abnormalities in Ca ++ regulation may lead to increased diastolic Ca ++ , which is the hallmark of contractile dysfunction in end-stage chronic HF patients. Therapeutic measures targeting Ca ++ homeostasis may facilitate the reverse remodeling in HF patients. Fischer et al. recently showed that increased SR-Ca ++ leak correlated with deteriorating LV function after LVAD implantation, and this SR-Ca ++ leak was significantly reduced by Ca ++ calmodulin-dependent protein kinase II (CaMKII) inhibition. CaMKII hyperphosphorylates RYR2, leading to disturbed diastolic closure of RYR2, and has a cascading effect on the Ca ++ homeostasis in HF patients. The observation that CaMKII inhibition reduces the SR-Ca ++ leak in HF suggests that CaMKII inhibition may be a promising therapeutic target for cardiac remodeling after LVAD implantation. Furthermore, CICR in rodent HF models suggests that the mechanical unloading by heterotopic abdominal heart transplantation increases calcium transient amplitude and normalizes the calcium spark frequencies and L-type calcium channel activity in cardiomyocytes. The role of short-term continuous-flow LVAD unloading was studied in an ovine model of acute myocardial infarction (MI), where it was observed that mechanical unloading prevented cardiac remodeling and dysfunction. Compared to a non-LVAD control group, calcium handling proteins were not altered, thereby preserving calcium cycling and improving cardiac function. These recent investigations demonstrate that extensive structural remodeling of the t-system, along with its depletion and change in orientation, plays an important role in the development and progression of ventricular dysfunction.
The effect of mechanical unloading on cytoskeletal proteins has been studied in LVAD clinical studies. Birks et al. showed significant differences (between the time of implantation and explantation/transplantation) in the regulation of nonsarcomeric proteins (lamin A/C, spectrin), integrins (β1, β5, β6, α5, and α7), and sarcomeric proteins that changed only in the recovered group (β-actin, α-tropomyosin, α1-actinin, and α-filament A). They also reported that vinculin, Wiskott-Aldrich syndrome protein, p21-activated kinase, Rho, and Graf levels decreased in the recovery group and increased in the nonrecovery group. These observations warrant further investigations on the role of the previously reported proteins in cardiac reverse remodeling. Cardiac ankyrin repeat protein (CARP) has been associated with cardiac hypertrophy, acts as transcription cofactor under mechanical stress and pressure overload, and has a critical role in the Nkx2.5 transcriptional pathway that regulates the ventricular muscle gene expression in the developing heart. A recent report on the CARP revealed increased levels in HF that eventually returned to normal levels after LVAD support. They also reported that CARP levels were increased after transaortic constriction in mice lacking βII spectrin, which the authors previously reported as an essential nodal protein following its significant alteration in their human and animal HF studies. Together, these observations indicate the potentially important role of cytoskeletal proteins in the myocardial reverse remodeling processes and further investigations are warranted.
Cardiac metabolism and bioenergetics
The heart utilizes multiple energy substrates like fatty acids, glucose, ketone bodies, lactate, and amino acids to meet its enormous energy demand for sustaining normal contractile function. This versatile use of substrates to meet its metabolic demand (i.e., the metabolic flexibility of the heart) is impaired during HF. This metabolic remodeling is integral to HF development and progression, with the failing heart reverting to a fetal pattern of energy substrate metabolism, i.e., enhanced glycolysis with decrease in fatty acid oxidation. This decrease in fatty acid metabolism seems beneficial in cardiac ischemic insults leading to hypoxic conditions, where anaerobic respiratory processes like glycolysis are beneficial, but may be deleterious in other cardiac conditions. Mechanical unloading with LVAD may reverse the maladaptive metabolic switches of the failing heart and even activate cellular pathways of cardioprotection and repair. Studies on LVAD-induced mechanical unloading have shown (1) improved respiratory capacity and augmented nitric oxide–dependent control of mitochondrial respiration ; (2) normalization of cardiolipin, a lipid component of the mitochondrial membrane important for adenosine triphosphate (ATP) formation and substrate transport, suggesting improved mitochondrial structure and function ; (3) depletion of long-chain acyl carnitine levels suggesting impairment of mitochondrial FA oxidation in failing hearts and substrate versatility of the mammalian heart ; (4) increased concentration of several amino acids, such as alanine, leucine/isoleucine, glutamine/glutamic acid and citrulline ; and (5) significant upsurge in the levels of total cholesterol, high-density lipoproteins, low-density lipoproteins, and triglycerides after LVAD implantation. Also, this increased level of total cholesterol from baseline to 3 months and a higher absolute cholesterol level postimplantation (at 3 months) were strongly associated with decreased mortality rate. From all of the aforementioned studies, it is evident that LVAD unloading relieves a significant part of the metabolic distress in the failing heart.
One intriguing concept is that glycolysis, although less efficient in terms of generating ATP, could be a sufficient energy source due to the decreased myocardial energy demand during unloading. Recent investigations on metabolic changes during HF showed upregulation of glycolysis without a subsequent increase in pyruvate oxidation through the Krebs cycle. This mismatch between glycolysis upregulation and mitochondrial pyruvate oxidations was attributed to the impaired structure and function of mitochondria ( Fig. 9.5 ). These metabolic changes may also be associated with the shift toward ketone body metabolism during chronic HF. The earlier observations, along with post-LVAD alterations in the expression of several metabolism-related genes and proteins, require further investigation to elucidate the specific role of cardiac metabolic adaptations during myocardial reverse remodeling.
Cell death and stress
Chronic HF is accompanied by loss of myocytes resulting in increased demand and stress. Cell death is mainly mediated by three processes, namely, apoptosis, necrosis, and autophagy. Apoptosis plays an important role in ventricular remodeling and contributes to the progressive impairment of contractile function. Apoptosis is mostly mediated through extrinsic (death receptors) and intrinsic (mitochondrial pathways) pathways resulting in cell shrinkage, fragmentation, and phagocytosis of the fragmented particles by the macrophages or neighboring cells. Several studies demonstrating changes compatible with reduced apoptosis during LVAD unloading were reviewed in detail by Soppa et al., who suggested that LVAD unloading interrupts apoptosis, and the increased levels of apoptosis in HF might be reversible via LVAD support. Correspondingly, a recent study showed that BCL2 associated X protein (Bax) and heat shock protein 72 (Hsp72) were significantly increased post-LVAD, suggesting a change in cardiac apoptotic environment and activation of antiapoptotic or protective factors. In contrast, patients following continuous-flow LVAD implantation showed a slight increase in caspase-1 signaling along with an escalation in inflammation and other apoptotic markers. Matrix metalloprotease-2 (MMP-2)/Janus kinase–mediated apoptotic and β1D-integrin/focal adhesion kinase–mediated survival pathways were reported to be activated in the nonischemic adjacent zone after MI in sheep, where LVAD unloading for 2 weeks attenuated remodeling, in part, by its negative effect on stretch-induced apoptosis and inhibition of MMP-2 activity. All in all, the majority of studies reported decreased levels of apoptosis upon unloading, which likely plays a role in the reverse remodeling and subsequent improvement in cardiac function and structure.
Markers of autophagy have been shown to be downregulated after LVAD unloading of failing hearts. Autophagy is a lysosome-mediated clean-up process, where damaged cytoplasmic organelles and proteins are digested through Beclin-1–initiated autophagocytosis to generate ATP. In cardiomyocytes, autophagy is also promoted via Fork-head box class O (FoxO) family transcription factors and plays a role in hypertrophy regression. Experimental studies have reported increased autophagy and levels of FoxO1 proteins after unloading. Atg5 is an essential molecule in the autophagosome assembly and is crucial for autophagy-mediated degradation. Atg5 -deficient mice showed significant decrease in cardiac contractility following mild transverse aortic constriction and angiotensin II treatment (known to induce hypertrophy). After the aortic constriction was removed (i.e., unloading), only controls showed signs of regression of hypertrophy, suggesting that autophagy also plays a role in the regression of LV hypertrophy. Lastly, reports on cell stress during HF and with LVAD-unloading showed decreased mitochondrial reactive oxygen species production and increased mitochondrial coupling efficiency. These favorable changes in myocyte attrition are complemented by data suggesting that LVAD unloading reduces myocardial stress, as indicated by reductions in the stress proteins metallothionein and hemeoxygenase-1. The aforementioned studies reveal that cardiomyocytes modify cell death pathways during overload and might use them to renew the damaged organelles when unloaded via LVAD.
Natriuretic peptides and neurohormones
Neurohormones drive survival and injury responses via several mechanisms. In chronic HF, neurohormonal responses are mostly triggered as a response to reduced myocardial contractility and hemodynamic overload, and levels of natriuretic peptides are used diagnostically. The sympathetic nervous system and the renin-angiotensin-aldosterone system play a crucial role in vasculature, cardiac, and renal hypertrophy and progression of HF. Studies investigating β-adrenergic signaling and sympathetic innervation reported that pulsatile flow LVAD unloading improved (1) β-adrenergic receptor density, location and distribution pattern, contractile response to β-adrenergic stimuli, and adenyl cyclase activity and (2) sympathetic innervation in the failing heart accompanied by clinical, functional, and hemodynamic improvements. Interestingly, continuous-flow LVADs are associated with higher muscle sympathetic nerve activity than pulsatile flow LVADs, suggesting that flow patterns and the degree of myocardial unloading may differ depending on the characteristics of a specific device.
Studies on pulsatile-LVAD unloading have been associated with decreased levels of atrial and brain natriuretic peptides (BNP) and tumor necrosis factor-α (TNF-α) in serum and myocardial tissue. Circulating levels of epinephrine, norepinephrine, renin, angiotensin II, and arginine vasopressin have been shown to decrease during pulsatile LVAD unloading, but the specific implications of these changes need to be further investigated.
Recent studies involving patients with continuous-flow LVAD reported decreased NT-proBNP (N-terminal fragment in the prohormone BNP) levels consistent with low ventricular wall stress. It was also proposed that postoperative improvement of BNP levels during LV unloading can be used as a predictive tool for postoperative outcomes such as length of stay, right ventricular failure, and postoperative survival. NT-proBNP levels decreased by 40%–60% following LVAD support, with incremental decreases to 80% with neurohormonal blockade drug therapy. Patients with lowest BNP levels 60 days postimplantation are more likely to experience functional recovery with the lowest risk of ventricular arrhythmia. In summary, most studies suggest a significant reduction in neurohormonal activation with LVAD support. Future investigations should study whether this is an epiphenomenon of other benefits derived by the MCS or an active mechanism that can be further manipulated to improve cardiac reverse remodeling and, in general, the cardiovascular health of mechanically assisted patients.
Inflammatory markers
Inflammation is one of the most crucial biological processes implicated in several aspects of cardiovascular disease. There is mounting evidence of the effect of LVAD support on the inflammatory response; however, some of the reports are conflicting and the complete picture of the effect of ventricular unloading on the HF inflammatory response is unclear. Significant increases in interleukin (IL)-6 and IL-8 have been identified in LVAD patient samples when compared to other advanced HF patients, while TNF-α remained unchanged. IL-33/suppression of tumorigenesis 2 (ST2) signaling in cardiac fibroblasts and cardiomyocytes is activated as a myocardial response to mechanical overload. A recent study showed that ST2 knockout mice exhibited increased LV hypertrophy, fibrosis, and apoptosis. Furthermore, recombinant IL-33 treatment following transverse aortic constriction reduced hypertrophy and fibrosis in wild-type mice, but not in ST2 knockout mice, suggesting that the IL-33/ST2 pathway plays an important role during mechanical overload. In addition, a recent study reported lower tissue expression of IL-33 and ST2 in LVAD patients compared to heart transplant patients. These investigators also observed that the levels of IL-33, ST2, IL-8, and TNF-α were low pre-LVAD and significantly increased after LVAD support. All together, these studies suggest that the IL-33/ST2 pathway is involved in the decline of cardiac function in HF and in reverse cardiac remodeling following LVAD support and therefore may serve as a new therapeutic target for myocardial recovery.
In another study, inflammatory markers before LVAD implantation and 3, 6, and 9 months following LVAD implantation were compared to those of a healthy control group. BNP levels and LV dimension by echocardiography dropped significantly after LVAD implantation ( Fig. 9.6 A ). Advanced HF patients undergoing LVAD implantation had elevated C reactive protein (CRP), monocyte chemoattractant protein-1, and interferon γ–induced protein 10 prior to LVAD implantation and remained elevated thereafter (see Fig. 9.6 B). However, levels of IL-8, TNF-α, macrophage inflammatory proteins-1β, and macrophage-derived chemokine significantly increased 9 months following LVAD implantation (see Fig. 9.6 C). These findings have been linked to LVAD-associated adverse events, and their full translational and clinical importance warrants further investigation.
Extracellular matrix and fibrosis
Investigations of the effect of LVAD unloading on extracellular matrix have shown conflicting results; a few studies reported decreased fibrosis, whereas other investigations found a significant increase in fibrosis. The explanation for the contradictory observations is not clear, with some attributing the inconsistent results to the differences in background medications, patient-specific heterogeneity, and the variable fibrosis assessment methodologies employed in these studies. A human myocardial tissue study addressed this controversial issue with the use of advanced image analysis techniques in whole-field digital microscopy, an approach that reduces observer bias, markedly increases the amount of myocardial tissue analyzed, and permits comprehensive endocardium-to-epicardium evaluation. Myocardial tissue from HF patients pre-LVAD implantation had increased interstitial and total fibrosis as compared to normal myocardium. The interstitial and total collagen content appeared to be further increased after pulsatile flow LVAD unloading in these patients. The effects of pulsatile flow LVAD unloading on the myocardial tissue levels of the renin-angiotensin-aldosterone axis and matrix metalloproteinases supported the previous results. Whether the observed increase in fibrosis is a manifestation of further progression of cardiac remodeling and HF (which LVAD support could not reverse) or a direct detrimental effect of MCS warrants further investigation.
In an attempt to identify novel biomarkers that regulate fibrosis during HF, Koshman et al. investigated the role of connective tissue growth factor (CTGF; a modulator of extracellular matrix production during inflammatory tissue injury) in the failing human heart and the response to mechanical unloading. While TGF-β1, CTGF, COL1-α1, and COL3-α1 mRNA levels were reduced by unloading, there was only a modest reduction in tissue fibrosis and no difference in protein-bound hydroxyproline concentration between pre- and post-LVAD tissue samples. They postulated that the persistent post-LVAD fibrosis may be related to a concomitant reduction in MMP-9 following unloading. They proposed that CTGF may be a key regulator of fibrosis during maladaptive remodeling and progression to HF, and although mechanical unloading normalizes most genotypic and functional abnormalities, its effect on extracellular matrix remodeling during HF appears to be incomplete. Along the same lines, Galectin-3 was identified as a regulator of myocardial fibrosis and was found to be elevated in end-stage HF patients. Following LVAD implantation, Galectin-3 levels were found to increase during the period of LVAD support. Lastly, Lok and colleagues found an increase in fibrosis as well as circulating profibrotic markers after continuous-flow LVAD support. The same group investigated the role of growth differentiation factor-15 (GDF-15; a stress-responsive cytokine) during LVAD support and found that high circulating levels of GDF-15 correlate with myocardial fibrosis that decline after 1 month of mechanical unloading and subsequently remain stable. Furthermore, they observed that the cardiac mRNA and protein expressions of GDF-15 are very low, suggesting that the heart is not an important source of GDF-15 production in these patients.
Several studies suggest that LVAD mechanical unloading is associated with altered extracellular matrix synthesis. Recently, persistent myocardial fibrosis (but no further increase) has been demonstrated in patients with end-stage HF supported with LVAD. Gene expression analysis on isolated cardiac macrophages revealed decreased Kruppel-like factor-4, transglutaminase-2, mannose receptor-1, and IL-10, along with reduced levels of metalloproteinase-2 and profibrotic M2 macrophage gene expression after LVAD unloading. These findings suggest persistence of inflammation-activated fibroblasts after LVAD unloading that might contribute to poor extracellular matrix turnover and impede the potential of myocardial recovery observed in a subset of these patients. The same investigators aimed to test the hypothesis that bone marrow stem cells (CD34 +) injected in the epicardium would improve histological measurements of vascularity, fibrosis, and inflammation in human subjects ( n = 6) undergoing LVAD placement. They found no significant differences in fibrosis or inflammation between groups; however, the density of activated fibroblasts was decreased in both CD34 +- and CD34-injected areas. These findings suggest that changes in myofibroblast numbers might be insufficient to alter myocardial fibrosis. In tandem, Sakamuri et al. found that mechanical unloading can restore several components of the fibrillar extracellular matrix and the basement membrane but at the same time showed persistent elevation of adverse remodeling of the nonfibrillar extracellular matrix, which could contribute to disease progression and HF relapse after removal of LVAD support.
Given the mixed results, further studies to determine the effects of MCS on extracellular matrix and identification of adjuvant therapies that could enhance the potential of myocardial recovery during LVAD support are needed.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
